Audio-Video Biometric System with Liveness Checks

Girija Chetty, Michael Wagner

Research output: A Conference proceeding or a Chapter in BookConference contributionpeer-review


In this paper we propose multimodal fusion of super resolved texture (SRT) features and 3D shape features
with acoustic features for 3D audio-video person authentication systems with liveness checks. The proposed SRT
features allow information related to non-rigid variations on speaking faces, such as expression lines, gestures,
and wrinkles, enhancing the performance of the system against impostor and spoof attacks. Experiments with
multimodal fusion of acoustic and super-resolved texture and 3D shape features for two different speaking face
data corpus, VidTIMIT, and AVOZES, allowed equal error rates (EERs) of less than 0.5 % for imposter and
type-1 replay attacks (still photo and pre-recorded audio) and less than 3% for more complex type-2 replay
attacks (pre-recorded video or fake CG animated video).
Original languageEnglish
Title of host publicationProceedings of the Conference on Image and Vision Computing - New Zealand
EditorsBrendan McCane
Place of PublicationNew Zealand
PublisherUniversity of Otago, Dunedin, New Zealand.
Number of pages6
ISBN (Print)00473105233
Publication statusPublished - 2005
EventIVCNZ-2005 - Dunedin, New Zealand
Duration: 28 Nov 200529 Nov 2005


Country/TerritoryNew Zealand


Dive into the research topics of 'Audio-Video Biometric System with Liveness Checks'. Together they form a unique fingerprint.

Cite this